Barker, P. R, Kennett, J. P., et al., 1990
Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 113
16. HYDROCARBONS IN SEDIMENT OF THE WEDDELL SEA, ANTARCTICA1
Keith A . Kvenvolden, 2 Frances D . Hostettler, 2 and Tamara J. Frank 3
ABSTRACT
Heavy hydrocarbons (~C 15 + ) are ubiquitous but minor components in sediment from three sites (692, 693, and
694) drilled during Ocean Drilling Program (ODP) Leg 113. This preliminary report is the first to describe the distribution of some of these hydrocarbons in sediment of the Weddell Sea. Samples from Sites 692 and 693, located on a midslope bench along the margin of Dronning Maud Land, span a time interval from Early Cretaceous to Pleistocene. In
our samples from the Tertiary portion of the record, having an average organic-carbon content of about 0.2%, «-alkanes are common and are characterized by populations that indicate both marine and terrigenous sources. The a/3-hopanes and aaa-steranes are minor components with isomer ratios indicating partial to full maturity. In contrast, our
samples from the Cretaceous portion of the record, having an average organic carbon content of about 4%, contain
mixtures of hydrocarbons in which /j-alkanes are secondary in abundance to the isoprenoid hydrocarbons, pristane and
phytane. Diasterenes, sterenes, and hopenes are present in anomalously high concentrations and indicate immaturity.
The Cretaceous hydrocarbons appear to be mainly primary, whereas the Tertiary hydrocarbons contain compounds that
indicate that the sediment, along with its organic content, has been recycled. Our samples from Site 694, located in the
Weddell Sea on the abyssal plain, range in age from late Miocene to early Pliocene, have an average organic-carbon content of about 0.2%, and containrt-alkaneswith a strong terrigenous signature. Isomer ratios of a/3-hopanes and aaasteranes indicate intermediate maturity; diasterenes, sterenes, and hopenes are absent. The hydrocarbons at this site are
mainly recycled and indicate erosion of previously buried sediment from the continent of Antarctica.
INTRODUCTION
During the last 5 years, interest in the heavy-hydrocarbon
(~Ci 5 + ) geochemistry of ocean sediment offshore from Antarctica has significantly increased. This increase is the result of
(1) a greater availability of samples due to more marine surveys
being undertaken by various nations in Antarctic waters, (2) an
awareness that the isolation of Antarctica provides an offshore
environment that has been relatively shielded from modern pollution, and (3) a desire by some to ascertain the petroleum potential of this vast offshore region. For example, in 1984 a marine geological and geophysical investigation was undertaken
offshore from Wilkes Land and in the Ross Sea (Fig. 1) by the
U.S. Geological Survey (Eittreim et al., 1984). Sediment samples from this investigation have been studied and the hydrocarbon geochemistry described (Kvenvolden et al., 1987a, b; Rapp
et al., 1987). These studies concluded that hydrocarbons came
from both marine and terrigenous sources and that the terrigenous component probably represents recycled sediment. Later
Venkatesan (1988a) described the lipid geochemistry, including
hydrocarbons, of some sediment samples collected in 1983 by a
team from New Zealand, and she also concluded that the organic matter is mainly recycled and mixed with some modern
marine material. Other related studies of Ross Sea sediment are
those of McIver (1975), who analyzed hydrocarbon gases from
Deep Sea Drilling Project sites, Sackett et al. (1974), who first
recognized the importance of organic matter recycling in Antarctic sediment, and Matsumoto et al. (1979), who studied a
dumpsite at McMurdo Station.
The Alfred Wegener Institute for Polar Research of Germany
conducted a marine geological survey in 1984 in the Bransfield
Strait, on the opposite side of the continent of Antarctica (Fig.
1). Geochemical studies of sediment samples from this area
1
Barker, P. R, Kennett, J. P., et al., 1990. Proc. ODP, Sci. Results, 113: College Station, TX (Ocean Drilling Program).
2
U.S. Geological Survey, Menlo Park, CA 94025.
3
Present address: Surface Science Laboratories, Mountain View, CA.
have indicated that (1) thermogenic hydrocarbons are present
locally (Whiticar et al., 1985), and (2) both marine and lesser
amounts of recycled terrigenous hydrocarbons are dispersed in
the sediment (Venkatesan and Kaplan, 1987).
In addition to these studies, Mackie et al. (1978) described
hydrocarbons in sediment from King Edward Cove on the island of South Georgia (Fig. 1) in the subantarctic region, and
Matsumoto et al. (1979, 1984) and Volkman et al. (1986) reported on hydrocarbons (attributed mainly to bacterial and algal sources) in lake water, sediment, and soil on the continent of
Antarctica. Also, Matsumoto et al. (1987) described steranes
and triterpanes from outcrop samples collected in southern Victoria Land (Fig. 1); their results show variable maturities reflecting the effects of the intrusion of basaltic dikes on sedimentary
organic matter.
With this background of hydrocarbon-geochemical information, Ocean Drilling Program (ODP) Leg 113 provided an unprecedented opportunity to extend investigations of hydrocarbons in sediment offshore from Antarctica to include the Weddell Sea (Fig. 1). Although nine sites were drilled on Leg 113, we
were able to obtain samples (Table 1) from only three sites (692,
693, and 694). Site 692 is located on a mid-slope bench on the
continental margin of Antarctica offshore from Dronning Maud
Land (Fig. 1) at a water depth of 2875 m. From this site we obtained two samples of Early Cretaceous, probably Berriasian to
Aptian or Albian age, organic-rich mudstone. The Early Cretaceous sediment at this site was encountered at a very shallow
sub-bottom depth of about 53 m. Much of the overlying Mesozoic and Cenozoic section is missing because of unconformities
and erosion by canyon cutting. About 30 km southwest of Site
692, and still on the mid-slope bench at a water depth of 2359
m, the stratigraphic section is more complete, and holes drilled
here (Site 693) cored sediment that ranges in age from Early
Cretaceous to Pleistocene. The Early Cretaceous sediment at
Site 693 is slightly younger (Aptian-Albian age) than the sediment found at Site 692. Unfortunately, no sample of this unit
from Site 693 was collected for us. Nevertheless, from this site
we obtained six samples of sediment ranging in age from early
Oligocene to Pleistocene (Table 1). Site 694 is located on the
199
K. A. KVENVOLDEN, E D. HOSTETTLER, T. J. FRANK
'
r,
60 °W
^
Vs*
/
511
/230
\
__
South
\ s
Georgia^T
/ ©
^v^/jp-a
30°E
C>°
30 °W
/
^
/
694 A ,'692 / 693
\ > o
\
.
\
/
60 °E
o
■3)
4s»x*
%
/
ANTARCTIC ' T l S }
PENINSULA.
i
90 °W
1
\
\
\
\
\
I
l
V^K J \
)
<* P ? \
l
I
|
1
l
/
\
/
3 ^
S^
South P o l ^ / ^
S/
» \
\
\
^K
s ^ \
\(
®£
^o
\
L
tf
/Ross
Ro
\
90 °E
I
/
'• \ X /% / *ss \ . |
%
*
s ^ \
/
'I
'/
l
/
>helfT -J
... \ <
• • > cc \
Mc Murdo | O \
\
st ation / {r
120°E
Antarctic
120°W
IO
*>>*>
/
^ — —
/
150°E
180°E
150°W
Figure 1. Map of Antarctica showing sampling Sites 692, 693, 694 from ODP Leg 113 in the Weddell Sea and regions where other organic-geochemical studies have been done, including DSDP Sites 330 and 511.
Table 1. Descriptions of sediment samples from the Weddell Sea, Antarctica.
Core,
Sample,
Number
Core
interval
(cm)
Sample
depth
(mbsf)
Water
depth
(m)
Organic
carbon
(*)
EOM
(Mg/g)
Pleistocene
e. Pliocene
e. Pliocene
e. Pliocene
1. Miocene
e. Oligocene
E. Cretaceous
E. Cretaceous
2359
2359
2359
2359
2359
2359
2875
2875
0.15
0.17
0.31
0.14
0.27
0.23
3.57
4.84
4.5
5.3
7.1
5.1
7.9
11
160
120
e. Pliocene
1. Miocene
1. Miocene
4653
4653
4653
0.34
0.15
0.16
4.5
0.4
3.4
Geologic
age
Dronning Maud Land margin — Sites 693 and 692
693A-4R-3
693A-9R-4
693A-12R-4
693A-18R-3
693A-25R-2
693A-40R-2
692B9R-2
692B-12R-1
129-132
125-128
125-128
125-128
129-132
143-146
60-63
118-121
26.2
75.9
104.7
161.2
227.4
371.8
70.9
89.4
Weddell Abyssal plain — Site 694
694B-3H-3
694B-24X-2
694C-6X-3
200
125-128
125-128
125-128
19.4
172.4
222.2
HYDROCARBONS IN SEDIMENT OF THE WEDDELL SEA
abyssal plain of the Weddell Sea, remote from continental areas
(Fig. 1). From this very deep-water site (4653 m), we obtained
three sediment samples ranging in age from late Miocene to
early Pliocene. These 11 samples provided material for us to
prepare this preliminary report, which is the first to describe the
hydrocarbon geochemistry of sediments of the Weddell Sea.
PROCEDURES
The 11 sediment samples described here were collected on
board ODP's JOIDES Resolution. An attempt was made to collect samples at approximately equal intervals down a hole, but
sampling could only be done in intervals of undisturbed sediment that contained enough material to provide for both sedimentological control and organic-geochemical analyses.
The geochemical procedures were described previously (Kvenvolden et al., 1987a, b) and are only summarized here. The sediment samples, collected shipboard, had been carefully wrapped
in cleaned aluminum foil and frozen immediately in plastic bags
for storage. When we were ready to analyze the samples, water
was removed by freeze-drying, and the samples were gently pulverized and sieved through a 32-mesh screen. We analyzed a
small portion of the sieved sediment for organic carbon by wet
combustion (Bush, 1970). For extraction and identification of
hydrocarbons, we used 50-75 g of pulverized and sieved sediment. Each sample was triple extracted with dichloromethane
by shaking on a wrist-action shaker. The combined filtered extracts were concentrated by rotary evaporation. Any sulfur was
removed by filtration through activated copper. We used a portion of the filtrate to determine the weight of the extract—the
extractable organic matter (EOM). The dichloromethane solvent was then exchanged for n-hexane.
The extracts in n-hexane were fractionated by liquid-solid
chromatography using a column packed with activated silica gel
and deactivated alumina. The column was sequentially eluted
with n-hexane, benzene, and methanol to yield fractions containing mainly aliphatic hydrocarbon, aromatic hydrocarbons, and
N, S, O-containing compounds, respectively. Only the aliphatic hydrocarbons were studied in detail by high resolution gas chromatography using a 30-m by 0.3-mm, DB-1 bonded phase,
fused-silica capillary column; aromatic hydrocarbons were analyzed only in selected samples. Identification of n-alkanes, isoprenoid hydrocarbons (pristane and phytane), and aromatic hydrocarbons was based on retention times and comparison with
standards. Relative proportions of the individual compounds
were determined by integrator peak heights. Portions of all hexane fractions and some benzene fractions were examined by gas
chromatography-mass spectrometry using a 30-m by 0.3-mm
DB-5 bonded phase, fused-silica column. Single-ion monitoring
(SIM) of m/z 191 and 217 produced mass fragmentograms from
which terpanes and steranes/diasteranes, respectively, were identified. In addition, SIM of m/z 57, 243, and 257 produced mass
fragmentograms of n-alkanes, fernenes, and diasterenes, respectively.
RESULTS
Geochemical results for 11 samples are shown in Tables 1
and 2; samples from proximal Sites 692 and 693 are arranged in
stratigraphic sequence, and the results from these sites are listed
separately from the results obtained at the remote Site 694. At
Site 693 the organic-carbon contents of the Cenozoic samples
are low (0.14%-0.31%) whereas the two Mesozoic (Early Cretaceous) samples (Site 692) are organic-rich (3.57% and 4.84%);
at Site 694 the organic-carbon values of the Cenozoic samples
are also low (0.15%-0.34%). Likewise, the EOM values follow
the same trend as the organic carbon with highest amounts
found in the two Early Cretaceous samples at Site 692 and much
lower amounts in the Cenozoic samples from Sites 693 and 694.
The aliphatic hydrocarbon fractions all contain n-alkanes
ranging from about n-C 13 to about n-C 36 . Representative chromatograms of the n-alkanes are shown in Figure 2; in addition,
pristane and phytane are shown along with other isoprenoid hydrocarbons. To illustrate the relations among n-alkanes and the
isoprenoid compounds pristane and phytane, various ratios are
listed in Table 2. The odd-even carbon number predominance
(OEPx) for the n-alkanes was calculated by the method of Scalan and Smith (1970); OEP17 and OEP 27 correspond to the n-alkane ranges C 1 7 ± 2 and C 2 7 ± 2 over which the calculations are
made. Values of OEP 17 are variable and, except for one sample,
show odd-carbon-number predominances. OEP 27 values usually
indicate large odd-carbon-number predominances for the highmolecular-weight n-alkanes. In the Early Cretaceous samples,
in which high-molecular-weight n-alkanes are partially masked
chromatographically by polycyclic-aliphatic hydrocarbons (Fig.
2D), relative proportions of individual compounds were estimated from chromatographic and mass spectral information;
these n-alkanes have an odd-carbon-number predominance (Table 2).
Pristane/phytane ratios increase with depth from 1.0 to 2.1
at Site 693 (Table 2); but at Site 692 the ratios of these compounds are about 1 (0.96 and 1.07), and they are the dominant
aliphatic hydrocarbons in the Early Cretaceous samples as indicated by average values of the ratios Pr/C 17 and Ph/C 18 (2.6 and
5.4, respectively). The isoprenoid hydrocarbons also become increasingly significant with depth relative to the n-alkanes at Site
694. The complexity of the relative distributions of aliphatic hydrocarbons—n-alkanes and pristane and phytane—at the three
sites is illustrated by representative chromatograms in Figure 2.
The differences in distribution patterns between Cenozoic (Site
693) and Mesozoic (Site 692) samples are dramatic (Fig. 2A-D).
Besides the identified n-alkanes and isoprenoid hydrocarbons, all
samples contain minor to significant branched and cyclic compounds. Particularly interesting are the high-molecular-weight aliphatics in the Early Cretaceous samples (Site 692), some of
which were identified by gas chromatography-mass spectrometry.
A portion of the aliphatic hydrocarbons in all samples is
composed of terpenoid and steroid compounds (Figs. 3 and 4).
At Site 693, the identified terpenoid compounds (Fig. 3A-C)
are principally a series of 17a(H),21/3 (H)-hopanes, including
Tm [17a(H)-22,29,30-trisnorhopane], and Ts [18a(H)-22,29,30trisnorneohopane]. Tm/Ts ratios are variable at this site; however, in three samples, Tm greatly exceeds Ts (Table 2). The
170(H), 2 la(H)-hopane series (moretanes) becomes increasingly
important with depth, and the 17/3(H),21/3(H)-hopanes are present
in only three of the six samples from this site (Table 2). Both 22S
and 22R epimers of 17a(H),21/3(H)-30-homohopane and of
17a(H),21;3 (H)-30,31-bishomohopane are present. The values
of the epimer ratios 22S/(22S + 22R) for three samples from
Site 693 (Table 2) are at or near the equilibrium value of full maturity of about 0.6 (Mackenzie, 1984), whereas in three samples
the ratios are not at equilibrium. At this site, steranes (Fig. 4AC) include the C27-, C28-, and C 29 -5a(H),14a(H),17a(H)-cholestane series with both the 20R and 20S isomers present. Only two
of the 20S/(20S + 20R) ratios of the C29a<ra sterane (Table 2)
are at the equilibrium value of 0.5 (MacKenzie, 1984). Individually, these compounds are present in lower concentrations than
the diasteranes, 13j3(H),17a(H)-diacholestane (20S) and 130(H),
17a(H)-diacholestane (20R) as illustrated in Figure 4A-C. Hopenes, sterenes, and diasterenes were not observed.
The terpenoid and steroid hydrocarbons in the Early Cretaceous samples at Site 692 (Figs. 3D and 4D) are in almost complete contrast to the compounds seen in the Cenozoic section at
Site 693. Although 17a(H),21/3 (H)-hopanes are present, they
are minor compared to 17jS(H),21/3 (H)-hopanes and the ho-
201
K. A. KVENVOLDEN, F. D. HOSTETTLER, T. J. FRANK
Table 2. Organic geochemical parameters in sediment samples from the Weddell Sea, Antarctica.
Sample
OEP17
OEP27
Pr
Ph
Pr
Cl7
m/z 217
m/z 191
Ratios of alkanes and isoDrenoids
Ph
Cl8
C 3 $
C
3°f£
C3iaP^
Tm
Ts
C
2 9
aaa^
Dominant
Steroid
Hydrocarbon
Dronning Maud Land margin:
693A-4R-3
693A-9R-4
693A-12R4
693A-18R-3
693A-25R-2
693A-40R-2
1.1
0.98
1.3
1.1
2.3
1.4
1.6
2.2
2.4
1.7
2.3
2.9
1.01
1.11
1.23
1.19
1.70
2.05
0.42
0.42
0.36
0.33
0.21
0.46
0.46
0.60
0.47
0.44
0.34
0.34
(XnoPP)
(Xnopp)
692B-9R-2
692B-12R-1
1.9
1.8
1.7
1.9
0.96
1.07
2.1
3.1
2.2
2.5
1.8
1.63
1.73
1.33
0.57
0.59
0.91
0.50
0.39
0.37
0.49
0.25
0.33
D
D
D
D
D
DeSt
0.64
0.38
12.2
10.6
2.6
9.4
2.1
2.9
0.59
0.56
0.35
0.55
0.38
0.39
0.85
0.86
1.6
0.91
3.6
6.6
4.4
6.4
1.6
1.3
1.5
1.5
0(no S)
0(no S)
nc(no Ts)
nc(no Ts)
0(noS)
0(noS)
St(no D)
St(no D)
0.70
0.85
1.6
(Kno PP)
(Kno pp)
(Kno PP)
2.2
7.3
5.7
.31
.47
.48
4.1
0.87
1.1
0.13
0.47
0.34
St
D
D
0.38
(XnoPP)
Weddell Abyssal plain:
694B-3H-3
694B-24X-2
694C-6X-3
1.4
1.5
1.8
Explanation: OEP, a measure of odd-even carbon number predominance (see text); Pr, pristane; Ph, phytane; CXj carbon number of molecule; 0$, Pa, PP,
hopanes; Tm and Ts, terpanes (see text); aaa, aaa-sterane; R and S, epimer configuration; m/z, mass to charge ratio of fragment ion; St, sterane; D,
diasterane; nc, not calculated.
penes. In addition, fernenes (m/z 243) were tentatively identified. The 22R-epimer of 17a(H),21j8 (H)-30-homohopane and
of 17a(H),21/3 (H)-30,31-bishomohopane greatly exceeds the
22S epimer. Tm is also present but not Ts. Only the 20R epimers
of
C27"> C2g-, anc * C29-5o:(H),14a!(H),17G!(H)-cholestane s e r ^ e s
and the 5/3(H),14a(H),17a(H)-cholestane series can be identified (Fig. 4D) along with possible 4-methyl steranes. The 130(H),
17a(H)-diacholestanes were not observed. Diasterenes (C 27 , C 28 ,
and C29) are also present as identified by single ion monitoring
of m/z 257 (Brassell et al., 1986); the 20R epimer is more abundant than the 20S epimer.
At Site 694 the dominant terpanes (Fig. 3E, F) are the
17a(H),21^(H)-hopane series and to a lesser extent the 17(8(H),
21a(H)-hopane series (moretanes). Both Tm and Ts are present
with Tm/Ts ratios ranging from 0.9 to 4.1 (Table 2). The
17|3(H),21a (H)-hopanes are not present in significant amounts.
The epimer ratios 22S/(22S + 22R) of the 17a(H),21/3(H)-homohopanes range from 0.31 to 0.48, indicating intermediate
maturity. Values of 0.13 to 0.47 for the epimer ratios 20S/(20S
+ 20R) of the C29-sterane [24-ethyl-5a(H), 14a(H),17a(H)-cholestane], the dominant sterane in the hydrocarbon mixtures,
also indicate intermediate maturity (Table 2). Diasteranes are
also present (Fig. 4E, F).
Identifiable aromatic hydrocarbons were found only in the
organic-rich Early Cretaceous mudstone at Site 692. In both
samples, perylene is the dominant aromatic compound, and it
constitutes at least 90% of the aromatic fraction of the deepest
sample. The remainder of this fraction contains a complex mixture of unidentified alkylated compounds. The aromatic fractions of the other samples from Sites 693 and 694 are composed
of very minor amounts of methylated and regular polycyclic aromatic hydrocarbons including traces of phenanthrene.
202
DISCUSSION
Our results show that n-alkanes, isoprenoid (pristane and
phytane), terpenoid, and steroid hydrocarbons are common,
but minor, components of Cenozoic sediment of the Weddell
Sea at least on the continental slope and abyssal plain. These
sediments are organic-poor, characterized by low amounts of
organic carbon and EOM, apparently resembling sediments offshore from Wilkes Land and in the Ross Sea (Fig. 1; see Kvenvolden et al., 1987b). In striking contrast are the results from
the Mesozoic sediment sampled from the continental slope; two
samples contain the same general classes of hydrocarbons but
different concentrations, distributions, and structural configurations. These sediments are organic-rich and, therefore, were
likely deposited under conditions where organic matter was abundant and preservation was efficient.
Cenozoic Section
Although the Cenozoic sediments from Sites 693 and 694
share the common characteristic of being organic-poor, their
hydrocarbon geochemistries reveal slight differences at the two
sites. For example, in most cases the distributions of aliphatic
hydrocarbons at Site 693 on the continental slope are more complex, having more unidentified components and relatively more
n-alkanes of lower molecular weights (Fig. 2A-C). Except for
one sample, n-C15 or n-C17 dominates the low-molecular-weight
mixtures, and the heavy n-alkanes show a variable odd-carbonnumber predominance (OEP 27 = 1.6-2.9). Pristane/phytane ratios are always greater than one, and, with one exception, n-C 17
and n-C18 are always more abundant than the accompanying
isoprenoid hydrocarbons. The n-alkane distributions of the abyssal plain sediments (Site 694) show two distinct modes (Fig. 2E,
HYDROCARBONS IN SEDIMENT OF THE WEDDELL SEA
693A-4R-3
r
Pr
692B-9R-2
(Adipate
artifact)
27
Cholestenes
Diacholestenes
oraa(20R )Cholestanes
Hopenes, Fernenes
170(H)-Hopanes
20
mi
\
1—
m\ *
JPw
JWW
^
I
Ill 33
35 37
hi ■^L~L^_JL_4
^ H k J L J U j MJWW
15
17
n
693A-18R-3
694B-3H-3
693A-40R-2
694C-6X-3
13
18
A
„
21
23
10
25
CO \
O \
Ph
i
/ /
B
25
27
31
25
Pr
Ph
\
Hi mk Uw JNMIMWUviU J4^ikii>+^
TIME(min)
UJIWJMJW* w
TIME(min)
Figure 2. Gas chromatograms of the aliphatic hydrocarbons in six sediment samples from the Weddell Sea, Antarctica. The «-alkanes are indicated by
carbon number; the carbon number of unnumbered peaks can be inferred from the carbon number of nearby homologs. Pr = pristane; Ph = phytane; and i = other isoprenoid hydrocarbons.
-F), one at «-C17 and the other at high molecular weights where
OEP 27 values of 1.8-2.5 indicate a significant odd-carbon-number dominance. Pristane and phytane become increasingly important with depth relative to n-C 17 and «-C18 (Table 2).
The lower molecular weight n-alkanes at both Sites 693 and
694 likely represent marine sources, an interpretation supported
by Clark and Blumer (1967), Han and Calvin (1969), and Simoneit (1978). The higher-molecular-weight n-alkanes with the
odd-carbon-number dominances likely represent contributions
of waxes from vascular plants of terrigenous (continental) origin (Eglinton et al., 1962). This interpretation is based on many
observations of marine sediments worldwide and summarized
by Simoneit (1978).
The cyclic biomarkers show contrasting maturities at the
Sites 693 and 694 with the sediments at Site 693 having both
mature and intermediate-maturity mixtures. At Site 693, the ho-
203
K. A. KVENVOLDEN, F. D. HOSTETTLER, T. J. FRANK
100
a/730
693A-4R-3
692B-9R-2
em
a/?31
Ts
fl '
a/?33
n
Tm
0a2Q
\0a3O
a 031
S R
JAJUL/W^
100
693A-18R-3
a/?20
694B-3H-3
LU
CO
z
o
co
LU
DC
B
100
a/?30
a £30
694C-6X-3
a/329
a£3L
n.
R
Ts
Tm
/?a30
T/J
a/J32
n
S R
0o2fl
^
T
> Vbs-^s-s
0
40
c
TIME(min)
80
40
F
TIME(min)
80
Figure 3. Mass fragmentograms (m/z 191) of terpenoid hydrocarbons in six sediment samples from the Weddell Sea, Antarctica. Carbon numbers are
indicated near peaks, a/3 = a/3-hopanes; 0/3 = /3(3-hopanes; /3a = /3a-hopanes (moretanes); a = A17(21)-hopene; b = A13(18)-neohopene; R and S =
C22 epimers; Ts and Tm, defined in text; T/3 = 17/3(H)-22,29,30-trisnorhopane.
panes are dominated by the highly stable a/3-hopane series. The
a/3-homohopane epimers in three samples are at equilibrium
(full maturity). Tm/Ts ratios range from 0.85 to 6.6, with the
largest value found in the sediment sampled at the greatest
depth below the seafloor (Tables 1, 2). The /3a-hopanes (moretanes) of intermediate stability increase in importance with depth.
The unstable /3/3-hopanes are present in three of the six samples.
Although the a|8-hopanes are dominant at Site 694, the /3a-hopanes (moretanes) are present in significant amounts. The afihomohopanes are only partially isomerized; that is, the epimers
are not at equilibrium. At Site 694, ratios of Tm/Ts vary from
204
0.9 to 4.1 (Table 2). No hopenes could be identified at this site.
Epimer ratios of the C29Q!Q!a-steranes provide further support
for the relative maturities at Sites 693 and 694 (Table 2). Thus
the cyclic-biomarker evidence, based on considerations of the
stability of structural isomers and the epimer ratios, clearly
demonstrates that Site 694 has intermediate-maturity mixtures,
similar to those observed in Ross Sea sediment (Kvenvolden et
al., 1987b), and that both intermediate and fully mature mixtures occur at Site 693.
Of course, the question arises as to how unconsolidated Cenozoic sediment, which has not been deeply buried and which
HYDROCARBONS IN SEDIMENT OF THE WEDDELL SEA
100
£«D29
692B-9R-2
0aa27
IK
693A-4R-3
100,-
^V/AJ
£c*D29
-£aD29
£aD27
KiUv^M
M
' l " .V'
V
694B-3H-3
TIME(min)
60
c 30
TIME(min)
Figure 4. Mass fragmentograms (m/z 217) of steroid hydrocarbons in six sediment samples from the Weddell Sea, Antarctica. Carbon numbers are
indicated near peaks, a a a = aaa-steranes; /3aa = (3aa steranes; /3aD = diasteranes; R and S = C2o epimers.
205
K. A. KVENVOLDEN, F. D. HOSTETTLER, T. J. FRANK
has not obviously been subjected to high temperatures, can contain some hydrocarbons signaling intermediate to full maturity.
The answer to this question was previously proposed in other investigations in the Antarctic (Sackett et al., 1974; Kvenvolden et
al., 1987b; Venkatesan, 1988a). Much of the sediment, containing hydrocarbons from both marine and terrigenous sources,
must have been recycled; that is, the sediment constituents must
be older than the Cenozoic time of sedimentation. Perhaps the
sediment had been eroded earlier from exposed rock that were
previously buried on the continent of Antarctica. Ice-rafted debris from multiple sources may have been involved. This scenario would help explain (1) the terrigenous signature of n-alkanes which may have come from vascular plants that were once
present on Antarctica but are now absent, and (2) the range of
maturities shown by the cyclic biomarkers.
Mesozoic Section
The hydrocarbon geochemistry of the two Early Cretaceous
(possibly Berriasian to Albian age) samples at Site 692 reveals a
paleohistory in which the organic-rich sediment has undergone
little diagenetic change since the time of deposition more than
130 million years ago. In general, our results are comparable to
the observation of Brassell et al. (1983), who pointed out that
deep-sea sediments of Cretaceous age commonly show little evidence of diagenetic change. Apparently these sediments have
never been deeply buried and have never been subjected to anomalous thermal stress.
The distributions of n-alkanes (Fig. 2D) show predominances of
odd-carbon-numbers in the low-molecular-weight region (OEP17 =
1.8 and 1.9). The source of these compounds is believed to be
mainly aquatic algae. This interpretation is supported by the extensive compilation by Clark and Blumer (1967) in which they
showed that the distributions of n-alkanes in marine phytoplankton comprises mainly n-C 14 through n-C20 with an oddcarbon-number dominance. For carbon numbers greater than
C25, the n-alkanes are partially masked on the gas chromatograms (Fig. 2D) by a complex mixture of polycyclic hydrocarbons; therefore, the distributions of n-alkanes in this complex
region were not determined chromatographically with great confidence, but SIM of m/z 57 showed an odd-carbon-number predominance suggesting some terrigenous input. Between n-alkanes identified as n-C 33 and n-C 34 (Fig. 2D) is a distinctive
compound whose retention time suggested diploptene, previously identified in sediment of the Bransfield Strait and the
Ross Sea by Venkatesan (1988b). However, mass spectra and
comparison with a known standard indicate that the compound
in our samples is 17/3(H),21/3(H)-30-homohopane; no diploptene was detected.
The isoprenoid hydrocarbons pristane and phytane are about
equally abundant and dominate the n-alkanes (Fig. 2D). The
very large Pr/C 17 and Ph/C 18 ratios (Table 2) emphasize this
dominance. Pristane/phytane ratios have been used to assess
paleoenvironments (Didyk et al., 1978), but the application has
restricted use (ten Haven et al., 1987), and Pr/Ph ratios near
one are difficult to interpret (Powell, 1988). The Mesozoic sediment samples, which contain these isoprenoid hydrocarbons
have organic carbon contents of 3.6% and 4.8%. This large
amount of organic matter strongly suggests that the paleoenvironment of these sediments was anoxic. Under such conditions,
a Pr/Ph ratio much less than one would be expected if the general model of Didyk et al. (1978) is accepted; however, we believe that our data brings the utility of this model into question
(see also ten Haven et al., 1987). Although the specific source
for these isoprenoid hydrocarbons is not known, they likely represent mainly marine sources such as algae and bacteria. We believe that our observed Pr/Ph ratios were established mainly by
the contributing sources and that the ratios have changed very
206
little during the course of time because these sediments have undergone minimal diagenesis.
The hopanoid and steroid hydrocarbons in our Early Cretaceous samples comprise mixtures that are diagenetically very
immature. Hopane isomers exemplify this interpretation. For
instance, /3|8-hopanes are least stable and have the stereochemistry
of their biological, bacterial precursors (Ourisson et al., 1984),
whereas, /fo-hopanes are of intermediate stability, and a/3-hopanes are most stable (Seifert and Moldowan, 1978). In our two
samples, the least stable /3/3-hopane isomers dominate (Fig. 3D).
The more-stable hopane isomer configurations are present but
are relatively unimportant. Ts, the more stable of the Tm-Ts
pair, is not present. The 22R epimer greatly exceeds the 22S epimer of the a/3-homohopane, showing that the extent of epimerization, which normally results in a mixture of 22R-22S, is very
small. In addition to the foregoing evidence of immaturity, the
presence of significant amounts of unsaturated hopenes (Fig.
3D) supports the interpretation of immaturity.
The steranes in these two Early Cretaceous samples compose
a very simple mixture (Fig. 4D) compared to the complex mixtures observed in the Cenozoic samples (Fig. 4A-C, E, F). The
20Raao;-steranes are the dominant steranes followed by the
20R/3o!Q!-steranes. The 20Rao;a-steranes have undergone little or
no isomerization because the 20Saa:a:-steranes were not detected. This lack of isomerization, the presence of the unstable
20R/3aQ!-steranes, and the fact that diasteranes were not observed provide additional evidence that these samples have undergone little diagenesis and are immature. The presence of
diasterenes also indicates immaturity.
Perylene is the major aromatic compound in these Early
Cretaceous sediments. Its source is uncertain, but the compound probably represents the very early diagenetic product
from a single source, possibly diatoms. Venkatesan (1988c) reviewed the possible sources of perylene in sediments and concluded that perylene is common in anoxic marine sediments and
that diatoms are major potential precursors. Our data appear to
support these conclusions. Also of interest is the fact that perylene is the dominant polycyclic aromatic hydrocarbon in sediment of the Bransfield Strait (Venkatesan and Kaplan, 1987),
which is near the Weddell Sea (Fig. 1).
The Mesozoic sediment at Site 692 represents a paleoenvironment that favored both the production and preservation of
organic material. The extent of this paleoenvironment is not
known, but the same or similar organic-rich facies are found on
the Falkland Plateau (Fig. 1) (Barker, Dalziel, et al., 1977;
Ludwig, Krasheninnikov, et al., 1983), which in the Early Cretaceous was close to Site 692 (Norton and Sclater, 1979; Lawver et
al., 1985). The molecular organic geochemistry of our samples
from Site 692 is similar in many aspects to the results obtained
on Early Cretaceous samples from the Deep Sea Drilling Project at Site 330 (Simoneit, 1980) and, in particular, Site 511 (von
der Dick et al., 1983) on the Falkland Plateau (Fig. 1). At Site
511 similarities include the dominance of pristane and phytane
relative to the n-alkanes, the striking abundance of unsaturated
terpenoid and steroid hydrocarbons, and the unusually high
amounts of 170(H), 2 l/3(H)-30-homohopane.
The depositional setting of these organic-rich sediments has
been separated by plate-tectonic movement so that similar facies
are now found near to the Antarctic continent as well as in the
region of the Falkland Plateau. Apparently, not much has happened diagenetically since these sediments became part of the
geologic record; hydrocarbon mixtures are immature, indicating
that the post-depositional diagenetic history has been uneventful although the tectonic history has involved major rifting.
More mature facies of such organic-rich sediments would be excellent sources of petroleum and may have developed elsewhere,
for example, through deeper burial. The petroleum potential of
HYDROCARBONS IN SEDIMENT OF THE WEDDELL SEA
Mesozoic sediment of the Falkland Plateau was recognized earlier by Comer and Littlejohn (1977). Our report shows that p o tential source beds of Mesozoic age are also located very close
to the Antarctic continent.
CONCLUSIONS
During this preliminary study of the hydrocarbon geochemistry of Weddell Sea sediment, we found immature mixtures of
hydrocarbons in organic-rich Mesozoic sediment a n d mature
mixtures of hydrocarbons in Cenozoic sediment. The Cenozoic
sediment apparently contains only a small a m o u n t of primary
marine organic material. Most of this kind of organic material
must have been lost through oxidation; what remains is mostly
secondary and recycled. The small a m o u n t s of organic matter
in the Cenozoic sediments originally may have come from the
erosion of older sedimentary rocks of the continent of A n t a r c tica. This source seems to explain much of the hydrocarbon
geochemistry of Cenozoic sediment not only of the Weddell Sea
but elsewhere offshore from Antarctica.
The organic material in the Mesozoic samples is mainly primary, and the hydrocarbons apparently have undergone little
change during the history of the sediment, assuming minimal
burial depths and minimal thermal stress. If any recycled organic matter is present in these samples, it is masked by the
a b u n d a n t primary organic material. The Mesozoic sediment
studied here represents a paleoenvironment that favored the
production and preservation of organic matter and where diagenetic processes have had little impact. More mature facies of
this sediment could be considered potential sources of petroleum.
ACKNOWLEDGMENTS
We thank J. B. Rapp for his laboratory expertise and support
and L. R. Bader a n d J. S. Mascardo for their technical a n d clerical assistance in the preparation of this manuscript.
REFERENCES
Barker, P. R, Dalziel, I.W.D., et al., 1977. Init. Repts. DSDP, 36: Washington (U.S. Govt. Printing Office).
Brassell, S. C , Howell, V. J., Gower, A. P., and Eglinton, G., 1983.
Lipid geochemistry of Cretaceous sediments recovered by the Deep
Sea Drilling Project. In Bjor<£y, M., et al. (Eds.), Advances in Organic Geochemistry 1981. Chichester (Wiley), 477-484.
Brassell, S. C , Eglinton, G., and Jia Mo, R, 1986. Biological marker
compounds as indicators of the depositional history of the Maoming oil shale. In Leythaeuser, D., and Rullkotter, J. (Eds.), Advances
in Organic Geochemistry 1985. Org. Geochem., 10:927-941.
Bush, P. R., 1970. A rapid method for the determination of carbonate
carbon and organic carbon. Chem. Geol., 6:59-62.
Clark, R. C , Jr., and Blumer, M., 1967. Distribution of ^-paraffins in
marine organisms and sediment. Limnol. Oceanogr., 12:79-87.
Comer, J. B., and Littlejohn, R., 1977. Content, composition, and thermal history of organic matter in Mesozoic sediments, Falkland Plateau: In Barker, P. R, Dalziel, I.W.D., et al. Init. Repts. DSDP, 36:
Washington (U.S. Govt. Printing Office), 941-944.
Didyk, B. M., Simoneit, B.R.T., Brassell, S. C , and Eglinton, G.,
1978. Organic geochemical indicators of paleoenvironmental conditions of sedimentation. Nature, 272:216-222.
Eglinton, G., Gonzales, A. G., Hamilton, R. J., and Raphael, R. A.,
1962. Hydrocarbon constituents of the wax coatings of plant leaves:
a taxonomic survey. Phytochemistry, 1:89-102.
Eittreim, S. L., Cooper, A. K., et al., 1984. Marine geological and geophysical investigations of the Antarctic continental margin, 1984.
U.S. Geol. Surv. Circ, 935:1-12.
Han, J., and Calvin, M., 1969. Hydrocarbon distribution of algae and
bacteria and microbial activity in sediments. Proc. Nath Acad. Sci.
(U.S.A.), 64:436-443.
Kvenvolden, K. A., Golan-Bac, M., and Rapp, J. B., 1987a. Hydrocarbon geochemistry of sediments offshore from Antarctica: Wilkes
Land continental margin. In Eittreim, S. L., and Hampton, M. A.
(Eds.), The Antarctic Continental Margin: Geology and Geophysics
of Offshore Wilkes Land. Circum-Pacific Council for Energy and
Mineral Resources Earth Science Series 5A:205-213.
Kvenvolden, K. A., Rapp, J. B., Golan-Bac, M., and Hostettler, F. D.,
1987b. Multiple sources of alkanes in Quaternary oceanic sediment
of Antarctica. Org. Geochem., 11:291-302.
Lawver, L. A., Sclater, J. G., and Meinke, L., 1985. Mesozoic and Cenozoic reconstructions of the South Atlantic. Tectonophysics, 114:
233-254.
Ludwig, W. J., Krasheninnikov, V. A., et al., 1983. Init. Repts. DSDP,
71: Washington (U.S. Govt. Printing Office).
Mackenzie, A. S., 1984. Application of biological markers in petroleum
geochemistry. In Brooks, J., and Welte, D. (Eds.) Advances in Petroleum Geochemistry, v. 1: London (Academic Press), 115-214.
Mackie, P. R., Platt, H. M., and Hardy, R., 1978. Hydrocarbons in the
marine environment, II. Distribution of /i-alkanes in the fauna and
environment of the sub-Antarctic island of South Georgia. Estuarine Coastal Mar. Sci., 6:301-313.
Matsumoto, G. I., Machihara, T., Sutuki, N., Funaki, M., and Watanuki, K., 1987. Steranes and triterpanes in the Beacon Supergroup
samples from southern Victoria Land in Antarctica. Geochim. Cosmochim. Acta, 51: 2663-2671.
Matsumoto, G. I., Torii, T., and Hanya, T , 1979. Distribution of organic constituents in lake waters and sediments of the McMurdo
Sound region in the Antarctic. Mem. Nat. Inst. Polar Res. Special
Issue, 13:103-120.
, 1984. A review of organic geochemistry in Antarctica. Mem.
Nat. Inst. Polar Res., Special Issue, 33:204-217.
McIver, R. D., 1975. Hydrocarbon gases in canned core samples from
Leg 28 Sites 271, 272, and 273, Ross Sea. In Hayes, D. E., Frakes,
L. A., et al., Init. Repts. DSDP, 28: Washington (U.S. Govt. Printing Office), 815-817.
Norton, L O . , and Sclater, J. G., 1979. A model for the evolution of the
Indian Ocean and the break-up of Gondwanaland. J. Geophys. Res.,
84: 6803-6830.
Ourisson, G., Albrecht, P., and Rohmer, M., 1984. The microbial origin
of fossil fuels. Sci. Am., 251:44-51.
Powell, T., 1988. Pristane/phytane ratio as environmental indicator. Nature, 330:641-643.
Rapp, J. B., Kvenvolden, K. A., and Golan-Bac, M., 1987. Hydrocarbon geochemistry of sediment offshore from Antarctica. In Cooper,
A. K., and Davey, R J. (Eds.), The Antarctic Continental Margin,
Geology and Geophysics of the Western Ross Sea. Circum-Pacific
Council for Energy and Mineral Resources Earth Science Series, 5B:
217-224.
Sackett, W. M., Poag, C. W., and Eadie, B. J., 1974. Kerogen recycling
in the Ross Sea, Antarctica. Science, 185:1045-1047.
Scalan, R. S., and Smith, J. E., 1970. An improved measure of the oddeven predominance in the normal alkanes of sediment extracts and
petroleum. Geochim. Cosmochim. Acta., 34:611-620.
Seifert, W K., and Moldowan, J. M., 1978. Applications of steranes,
terpanes and monoaromatics to the maturation, migration and source
of crude oils. Geochim. Cosmochim. Acta., 42:77-95.
Simoneit, B.R.T., 1978. The organic chemistry of marine sediments. In
Riley, J. P., and Chester, R. (Eds.), Chemical Oceanography 2nd ed.,
v. 7: New York (Academic Press), 233-311.
, 1980. Organic geochemistry of Mesozoic sediments from Deep
Sea Drilling Project Site 330, Falkland Plateau, In Lancelot, Y.,
Winterer, E. L., et al., Init. Repts. DSDP, 50: Washington (U.S.
Govt. Printing Office), 637-641.
ten Haven, H. L., de Leeuw, J. W , Rullkotter, J., and Sinninghe
Damste, J. S., 1987. Restricted utility of the pristane/phytane ratio
as a paleoenvironmental indicator. Nature, 330:641-643.
Venkatesan, M. I., 1988a. Organic geochemistry of marine sediments in
Antarctic region: marine lipids in McMurdo Sound. Org. Geochem.,
12:13-27.
, 1988b. Diploptene in Antarctic sediment. Geochim. Cosmochim. Acta, 52:217-222.
, 1988c. Occurrence and possible sources of perylene in marine sediments-a review. Mar. Chem., 25:1-27.
Venkatesan, M. I., and Kaplan, I. R., 1987. The lipid geochemistry of
Antarctic marine sediments: Bransfield Strait. Mar. Chem., 21:347375.
Volkman, J. K., Allen, D. I., Stevenson, P. O., and Burton, H. R.,
1986. Bacterial and algal hydrocarbons in sediments from a saline
207
K. A. KVENVOLDEN, F. D. HOSTETTLER, T. J. FRANK
Antarctic lake, Ace Lake. In Leythaeuser, D., and Rullkotter, J.
(Eds.), Advances in Organic Geochemistry 1985. Org. Geochem.,
10:671-681.
von der Dick, H., Rullkotter, J., and Welte, D. H., 1983. Content, type,
and thermal evolution of organic matter in sediments from the eastern Falkland Plateau, Deep Sea Drilling Project, Leg 71. In Ludwig,
W. J., and Krasheninnikov, V. A., et al., Init. Repts. DSDP, 71:
Washington (U.S. Govt. Printing Office), 1015-1032.
208
Whiticar, M. J., Suess, E., and Wehner, H., 1985. Thermogenic hydrocarbons in surface sediments of the Bransfield Strait, Antarctic Peninsula. Nature, 314:87-90.
Date of initial receipt: 30 January 1989
Date of acceptance: 30 June 1989
Ms 113B-163